It was no more than 20 years ago that biologists believed that cell adhesion molecules were simply the glue of life, the stuff that served to hold cells and ligaments and everything together. Since then, however, understanding of these molecules has gone through a paradigm shift. It is now known that they play roles in just about every aspect of human biology--from the embryo, where they are crucial for tissue and organ development, to the adult, where they act as traffic signals to direct the actions of immune-system cells in wound-healing, inflammation, cancer, and even AIDS.
Today the study of cell adhesion molecules has been transformed from a backwater of biology into one of the hottest fields around. "There are probably more immunologists working on adhesive molecules," says Harvard Medical School biologist Timothy A. Springer, "than there are on T cell receptors." Springer should know. His laboratory is responsible for defining many of the cell adhesion molecules now known, and Springer himself came in at #12 in the Science Watch ranking of the highest-impact biomedical scientists between 1990 and 1996. Springer published eight "blockbuster" papers during that period, and has had a total of almost 90 papers with more than 100 citations apiece. Four of his papers published since 1986 have been cited over a thousand times each, including his ultra-hot Nature paper, "Adhesion receptors of the immune system," which by last June had been cited over 3,600 times (A. Springer, Nature, 346:425-34, 1990).
Springer, 50, did his undergraduate work at the University of California at Berkeley, where he majored in biochemistry, and then went on to get a Ph.D. in 1976 at Harvard University in biochemistry and molecular biology. He then spent a year at Cambridge University working in the laboratory of the Nobel laureate Cesar Milstein. It was with Milstein, he says, that he first made monoclonal antibodies and realized their "enormous potential for dissecting molecules on the cell surface." In 1977, Springer joined the faculty at Harvard Medical School, where he is now professor of pathology at the Center for Blood Research.
Springer spoke to Science Watch from his office at Harvard.
SW: Let's begin with the obvious question. Your 1990 Nature paper has been cited an astronomical number of times. Why?
Springer: Aside from simply summarizing work in the field and providing a useful reference, I did synthesize a lot of information. One aspect was classifying adhesion molecules in different families: the immunoglobulin family, the integrin family, and the selectin family. I also talked about their different functional roles--for example, their ability to mediate signaling. And I discussed the relative sizes of these molecules, and how far apart cells were likely to be when they interacted through them. Adhesion was very hot in immunology at the time, so the paper received a lot of attention.
SW: Can you give us an overview of the different functions and families of cell adhesion molecules?
Springer: They fall into at least two different classes, and it's not a clean division. There are the signaling molecules, and the adhesive molecules--and in some cases you have adhesive molecules that also signal. It's not so simple. Some molecules are certainly very important for mediating adhesion. Take, for example, leukocyte interactions with endothelial cells in the bloodstream. There you have leukocytes binding to endothelium, and the flow of the blood actually exerts a very strong force on the adherent white blood cells, threatening to tear them away from the adhesive contact with the endothelium. So the molecules must be able to resist force. I would put the integrins, the selectins and the cadherins in the group that resists force.
Selectins and integrins can act in the vasculature. Integrins and cadherins are important throughout the body, probably for maintaining the integrity of tissues--holding all the tissues together. Then there's another class of molecules that are important in signaling, like CD2, LFA-3 or CD4 and CD8. And now we know that there's a much larger group of molecules that are important in the immune system, acting as receptor-ligand pairs. So you have FAS ligand; you have members of the TNF receptor family that recognize cell surface ligands. You have multiple immunoglobulin family molecules like CD28, and B7, which are important in co-stimulation. And you have CD40 on B cells that binds to a CD40 ligand on T cells, which is also important in co-stimulation. A lot of these molecules are delivering a signal, telling the cell that they're interacting with the right kind of other cell. Often that signal doesn't do anything unless it's received together with some other signals--for example, one coming from the T cell receptor.
SW: Can you give us an example of how this signaling works in the immune system?
Springer: It had been observed in vivo, for instance, that leukocytes in the circulation come along to a site of inflammation and bind to the vessel wall, at which point the first thing they do is roll along the wall in this very odd kind of transient adhesive interaction. So they have to be bound through adhesion molecules to the vessel wall. But the funny thing is that the zone of adhesive contact is translated along the vessel wall. It actually provides a mechanism for the white blood cell to survey the vessel lining for more signals. Next, the cells develop firm adhesion, so they stop rolling. If you make a video of this, you can even see them change shape and flatten out when they stop.
In 1990, Mike Lawrence and I showed in vivo that if we took purified selectins and put them on the wall of a flow chamber, leukocytes would come in and bind, and then they would start actually rolling along--just like in vivo. They did not pay attention to an immunoglobulin superfamily member, ICAM-1, on the wall along with the selectin. But we showed that if a chemoattractant is added to the solution being pumped through this chamber, within seconds you can see the cells stop rolling as they bind firmly to ICAM-1 on the substrate and then spread out. So we were able to recapitulate the steps in vivo that had been seen in vivo, showing that they were mediated by specific receptor-ligand pairs.
SW: Is this multi-step signaling conserved with many different cellular signals?
Springer: It turns out that it's also seen in lymphocytes, eosinophils, and monocytes, and it's been seen in vivo with endothelial monolayers as well as purified molecules on the substrate. So it seems quite general that selectins mediate rolling, and integrins can mediate firm adhesion--although certain integrins can mediate both rolling and firm adhesion, so they're sort of intermediate between the two.
We and others have extended it further and shown that one selectin molecule, L-selectin, can mediate binding between leukocytes and the endothelium, or can mediate leukocyte-leukocyte interactions. Other molecules, such as P-selectin and E-selectin, seem to mediate slower rolling. And then you have VCAM and MAdCAM, which mediate slower rolling and can also mediate firm adhesion, and so on.
The steps involved in leukocyte-endothelial binding and trans-endothelial migration are pretty well worked out. What's currently missing is an appreciation of how cells go from one location in a tissue to another--say, how T cells know to go to a T cell zone; and how B cells know to go to a B cell zone in lymphoid tissues.
SW: Back in 1992 there was a flurry of press reports about the biotech industry taking on cell adhesion molecules. What sparked the flurry, and has the situation cooled since then?
Springer: Part of what was being followed at that time was the RGD peptides, which are short peptides that are the key recognition motif for many integrins. Also, some work on rhinovirus had come out. It turns out that ICAM-1, the cell adhesion molecule that I've mentioned, is a receptor for rhinovirus. It looked like an interesting target for blocking replication of the cold virus. And the excitement hasn't fallen off. There's still a lot of interest in the pharmaceutical industry in developing these as targets. There's actually one that's a drug now--a monoclonal antibody to a platelet integrin. It blocks platelets from accumulating at sites of clotting after a myocardial infarction. It's been approved by the FDA.
SW: What would you now consider the most important clinical targets for the research on cell adhesion molecules?
Springer: There are a number of companies that have antibodies, or even small-molecule antagonists, in clinical trials now. These include antibodies against integrins, for treatment of graft rejection, and for prevention of traumatic stroke. Antagonists of selectins have been in clinical trials for a condition in which clots occur in the lung. A number of companies have been formed around the idea of adhesion. I founded a company called LeukoSite that has monoclonal antibodies in clinical trials, and we're developing antagonists for chemoattractants, with the idea of blocking leukocyte recruitment. In general, the antibodies against integrins could be used either acutely--say, to prevent shock or stroke, or to reverse some of the damage in stroke that is leukocyte-mediated. They might be used in chronic conditions such as graft rejection, or in diseases where there is a flare-up that needs to be treated promptly. We're also interested in inflammatory bowel disease. One can certainly think of targeting other kinds of autoimmune diseases, such as arthritis. The chemokine antagonists could also be used in diseases like rheumatoid arthritis or asthma--something with a chronic immune component. And, of course, there was a newspaper article last fall on the first results of a clinical trial of ICAM-1 for treatment of the common cold.
SW: Why a newspaper article and not a journal article?
Springer: Because it was presented as an abstract at a meeting. They gave it to patients who were dosed with rhinovirus--I guess these were college students. You lock them up and see what happens. So they gave ICAM-1 either at the same time as the rhinovirus or six hours after giving the rhinovirus, and it definitely reduced symptoms but didn't completely prevent infection. The patients reported that they felt better compared to the controls. The researchers used an objective grading system, called nasal mucus weight, and the ICAM-1 seemed to reduce nasal mucus weight by 50%.
SW: What do you see as the big open questions in the field as of early 1998?
Springer: One of the key questions now is how adhesion molecules function in signaling, and how they must coordinate with the cytoskeleton in order to do that. Many of the key players in that interaction are still not known. I'm personally working on the structure of these molecules, and I'm very interested in what we call inside-out signaling through integrins. That's when the cell rapidly changes the adhesiveness of the integrin. It's a very rapid, very dynamic way of regulating the adhesiveness of a cell. The other way would be to, for example, change the density of the molecule at the cell surface--obviously a slower and less dynamic way to do it. I'm interested in the structure of integrins and the conformational changes that regulate this. Another question: Now that we have really defined how leukocytes leave the vasculature, we have to figure out what determines where they go afterward. They take very specific pathways through tissues, but what signals that? What adhesion molecules do leukocytes use as they migrate through tissues? What are the guidance cues--are they the chemoattractants and their receptors? There is a lot of work being done right now on discovering new chemoattractants and new chemoattractant receptors.
SW: Lately there's been a lot of interest in cell adhesion molecules and AIDS. Could you explain the connection?
Springer: There is a set of traffic signals that helps direct white blood cells to the correct place in the body--leaving the blood stream and proceeding, for example, to the site of an inflammation. Some of these traffic signals are adhesion molecules; some are chemoattractant receptors. A chemoattractant receptor on T cells is identical to a receptor known as fusin, and the chemoattractant binding to that receptor blocks the binding of HIV and blocks infection. That's a real hot target now in the HIV field. It turns out there's another fusin, called CCR-5, and there are people who are genetically deficient in CCR-5 who don't get HIV. And now there are pharmaceutical companies trying to develop antagonists to block HIV infection using the receptor.
SW: Now that the competition in research on cell adhesion molecules has become so heated, has it changed the way you work?
Springer: Yes. One way I've changed my approach is to try to go off and discover some new aspect of this area. That's one reason I started working on chemoattractants and their receptors recently. The field has lately become very crowded and competitive.
High-Impact Papers from Timothy A. Springer Published Since 1990
(Ranked by average citations per year)
|Rank||Paper||Total Citations||per year|
|1||T.A. Springer, "Adhesion receptors of the immune system," Nature, 346(6283):425-34, 1990.||3,676||490|
|2||T.A. Springer, "Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm," Cell, 76(2):301-14, 1994.||1,088||311|
|3||M.B. Lawrence, T.A. Springer, "Leukocytes roll on a selectin at physiological flow rates: Distinction from and prerequisite for adhesion through integrins," Cell, 65(5):859-73, 1991.||810||125|
|4||M.S. Diamond, et al., ICAM-1 (Cd54): A counter-receptor for Mac-1 (Cd11B CD18," J. Cell l. Biol., 111(6):3129-39, 1990.||388||60|
|5||T.A. Springer, "Cell adhesion: Sticky sugars for selectins," Nature, 349(6306):196-7, 1991.||336||52|